In the process of making semiconductor devices a photoresist and an antireflective material are applied to a substrate. Photoresists are photosensitive films used to transfer an image to a substrate. A photoresist is formed on a substrate and then exposed to a radiation source through a photomask (reticle). Exposure to the radiation provides a photochemical transformation of the photoresist, thus transferring the pattern of the photomask to the photoresist. The photoresist is then developed to provide a relief image that permits selective processing of the substrate.
Photoresists are typically used in the manufacture of semiconductors to create features such as vias, trenches or combination of the two, in a dielectric material. In such a process, the reflection of radiation during exposure of the photoresist can limit the resolution of the image patterned in the photoresist due to reflections from the material beneath the photoresist. Reflection of radiation from the substrate/photoresist interface can also produce variations in the radiation intensity during exposure, resulting in non-uniform linewidths. Reflections also result in unwanted scattering of radiation exposing regions of the photoresist not intended, which again results in linewidth variation. The amount of scattering and reflection will vary from one region of the substrate to another resulting in further non-uniform linewidths.
With recent trends towards high-density semiconductor devices, there is a movement in the industry to use low wavelength radiation sources into the deep ultraviolet (DUV) light (300 nm or less) for imaging a photoresist, e.g., KrF excimer laser light (248 nm), ArF excimer laser light (193 nm), electron beams and soft x-rays. However, the use of low wavelength radiation often results in increased reflections from the upper resist surface as well as the surface of the underlying substrate.
Substrate reflections at ultraviolet and deep ultraviolet wavelengths are notorious for producing standing wave effects and resist notching which severely limit critical dimension (CD) control. Notching results from substrate topography and non-uniform substrate reflectivity which causes local variations in exposure energy on the resist. Standing waves are thin film interference or periodic variations of light intensity through the resist thickness. These light variations are introduced because planarization of the resist presents a different thickness through the underlying topography. Thin film interference plays a dominant role in CD control of single material photoresist processes, causing large changes in the effective exposure dose due to a tiny change in the optical phase. Thin film interference effects are described in “Optimization of optical properties of resist processes” (T. Brunner, SPIE 10 Proceedings Vol. 1466, 1991, 297).
Bottom anti-reflective coatings (BARCs) have been used with single material resist systems to reduce thin film interference with some success. However, BARCs do not provide control of topographic variations and do not address the differences in resist thickness. BARCs such as silicon nitride or silicon oxynitride typically follow the already existing topography, and thus, the BARC exhibits nearly the same thickness non-uniformity as the underlying material. Consequently, the BARC alone will generally not planarize topographic variations resulting from underlying device features. As a result, there will be a variation in exposure energy over the resist. Current trends to provide uniform topography via chemical/mechanical polishing still leaves significant variations in film thickness.
Variations in substrate topography also limits resolution and can affect the uniformity of photoresist development because the impinging radiation scatters or reflects in uncontrollable directions. As substrate topography becomes more complex with more complex circuit designs, the effects of reflected radiation becomes even more critical. For example, metal interconnects used on many microelectronic substrates are particularly problematic due to their topography and regions of high reflectivity.
One approach to variations in substrate topography is described in U.S. Pat. No. 4,557,797 (Fuller et al.). Another approach used to address variations in substrate topography is described in Adams et al., Planarizing AR for DUV Lithography, Microlithography 1999: Advances in Resist Technology and Processing XVI, Proceedings of SPIE, vol. 3678, part 2, pp 849-856, 1999, which discloses the use of a planarizing antireflective coating.
Although multimaterial patterning schemes exist in the prior art (see, U.S. Pat. No. 6,140,226; and R. D. Goldblett, et.al. Proceedings of the IEEE 2000 International Technology Conference, p 261-263), there remains the need for new antireflective materials. Many of the prior antireflective materials contain silicon based intermediate materials that do not act as antireflective coatings, e.g. silicon oxide like materials require the use of an additional antireflective coating because they cannot be optically tuned to control reflections.
To obtain optimal lithographic performance, the use of a relatively thin photoresist is desired. A thin photoresist provides for enhanced resolution and greater depth of focus. A thin photoresist, however, does not provide the etch performance needed for the transfer of the resist pattern into the underlying substrate. Significant resist loss occurs while the BARC is being patterned, in addition to striations and surface roughening. This problem becomes more serious if a thin photoresist is used.
It is desirable to develop a thin resist process which provides excellent lithographic performance and provides appropriate etch resistance for effective pattern transfer into the underlying substrate. To accomplish this, new antireflective materials are needed which provide better etch selectivity to resist than current silicon-based or organic antireflective materials. The antireflective materials need to (1) have appropriate optical properties at appropriate wavelengths, (2) provide good etch selectivity to resist (greater than 1:1) and (3) exhibit little, if any, interaction with the resist to minimize residue, footing and undercut which limits the overall lithographic performance of the resist.
The present trend to 248 nm and 193 nm lithography and the demand for sub 200 nm features also requires that new processing schemes be developed. To accomplish this, tools with higher numerical aperture (NA) are emerging. The higher NA allows for improved resolution but reduces the depth of focus of aerial images projected onto the resist. Because of the reduced depth of focus, a thinner resist is typically required. However, as the thickness of the resist is decreased, the resist becomes less effective as a mask for subsequent dry etch image transfer to the underlying substrate. In many instances, if a substantial etching depth is required, or if it is desired to use certain etchants for a given underlying material, the resist thickness is now insufficient to complete the etch process. Without significant improvement in the etch resistance exhibited by current single material resists, these systems cannot provide the necessary etch characteristics for high resolution lithography.